The Court of Appeal (in Graves v Brouwer) has taken the opportunity to re-visit the vexed question of how you should go about proving causation where there are a series of possible causes and, perhaps more importantly, how you shouldn’t do it. Anyone faced with a fire claim would do well to take heed.
The problem is easily stated, but not so easy to answer:
- Say there are four possible causes of a loss with, respectively, a 20%, 20%, 20% and 40% likelihood that they were the true cause. Is it right to say that the 40% cause, being twice as likely as any of the others, is, on the balance of probabilities, the cause of the loss?
- What if all the possible causes identified have a very small likelihood, but one is still significantly more likely than the others?
The courts have struggled to provide a consistent answer to this conundrum, but I think the position is now much clearer than it has been for decades.
Sherlock Holmes advises Dr Watson
For a while it appeared that the House of Lords’ case of Rhesa Shipping Co v Edmunds (The Popi M) had resolved the matter conclusively. Lord Brandon referred to Sherlock Holmes’ famous advice to Dr Watson:
“…when you have eliminated the impossible, whatever remains, however improbable, must be the truth.”
Lord Brandon went on to explain why this advice was wrong. There is a third alternative open to the court as it can find that the case is not proved. Also, the dictum is only logical if all relevant facts are known, which they rarely are.
Lord Brandon stressed that a court can only find for the claimant when its case is proved on the balance of probabilities and that cannot be a sensible conclusion where, as in The Popi M, a judge described a cause as “extremely improbable”.
However, although the more recent Court of Appeal case of Nulty v Milton Keynes firmly backed The Popi M (see also Akenhead J in Fosse Motor Engineers Ltd v Conde Nast), other Court of Appeal decisions have been less sure:
- In Kiani v Land Rover, Sherlock Holmes, it seemed, was able to come back from the dead, as Waller LJ accepted that:
“…where only two possibilities are under consideration, both of which seem unlikely, if one seems much less likely than the other the less likely can be discounted thus making the first likely to happen on the balance of probabilities.”
- In Ide v ATB, Sherlock was back working full time in Baker Street, as Thomas LJ accepted that:
“…as a matter of principle, if there were only three possible causes of an event, then it was permissible for a judge to approach the matter by analysing each of those causes. If he ranked those causes in terms of probability and concluded that one was more probable than the others, then provided those were the only three possible causes, he was entitled to conclude that the one he considered most probable, was the probable cause of the event provided it was not improbable.”
Graves v Brouwer
It is fair to say that the facts of Graves v Brouwer are unlikely to trouble the courts again in the near future, but the principles applied by the court most certainly will. As Tomlinson LJ conceded, it was “by any standards a bizarre case”.
Miss Graves and Mr Brouwer were next-door neighbours in Sutton. Miss Graves’ house was damaged by fire. It transpired that half an hour beforehand, Mr Brouwer had set fire to two A4 pieces of paper and two A4 pieces of card in a narrow alleyway between the two houses. The paper was two letters from the bank providing his new PIN numbers and the card was the back of notepads on which he had scribbled various facts and figures of a confidential nature. He had no shredder and so he burnt them (!).
Perhaps unsurprisingly, Miss Graves said that the fire in her house was caused by an escape of the fire lit by Mr Brouwer, and she alleged negligence. Mr Brouwer said that the fire in the house was not related to his small paper fire and, in any event, he denied negligence.
At first instance, the judge found that Mr Brouwer’s fire had indeed caused the fire in Miss Graves’ house, but that Mr Brouwer had not been negligent (he doused it with a hose). Miss Graves appealed and Mr Brouwer cross-appealed the respective decisions.
Graves v Brouwer at first instance
The competing theories for what caused the fire were:
- Floating embers from the alleyway fire (referred to as “flying brands”).
- Birds bringing in combustible material.
- Electrical defects.
Both experts agreed that the flying brands theory was of “a low order” of likelihood, but Miss Graves’ expert said it was still the most likely cause. Mr Brouwer’s expert thought arson more likely, but the judge found the following exchange in cross examination to be crucial:
Judge: “On the balance of probabilities, having eliminated all the evidence either of being of no account at all or your arson theory, it is, on the balance of probabilities, the most likely, is it not?”
Mr Brouwer’s expert: “I understand that, if the court deems that arson or deliberate fire setting in any manner is unlikely or less implausible than the fire from the items lit in the alley, then that would be the decision.”
Graves v Brouwer in the Court of Appeal
In short, the Court of Appeal fell solidly behind Nulty and The Popi M. This was a case which epitomised the problem with the “Sherlock Holmes approach”. The investigation into the fire had been inadequate. There had been no proper consideration or investigation of the other evident possibilities: arson, smoking materials, faulty cooking or heating appliances, bird nesting. The flying brands theory was accepted as scientifically improbable.
The line of cross examination which led to, what the judge considered to be a significant concession, was put upon a false premise. It presumed that there were only two possible causes, which was not the case. Consequently, the judge fell into error because:
“She did not ask herself whether the case for believing that the fire was caused in this way was stronger than the case for not coming to that belief.”
Where does that leave us?
Causation and probabilities can be difficult and counter-intuitive. It seems that, in order to ensure an “appeal-proof” decision, it is crucial that, whatever other convoluted analysis a judge might apply, at some point they need to stand back and assess whether the cause in question was or was not the likely cause of the loss on the balance of probabilities. In real terms, it is likely that Sherlock’s advice plays a part in that process, but the judgment, and therefore the evidence and submissions, must be careful not to make that the decisive rationale.
There is also a lesson for defendants. If you can get your expert to rank the possible causes and entice the judge down that road, you might be able to derail what would otherwise have been a perfectly valid finding on causation.
The two Court of Appeal decisions (Nulty and Graves) mark a clear statement of where the law now is. It is noticeable that in both cases the Court of Appeal took the trouble to explain the law in detail, overturn a first instance decision on the law, but all in circumstances that made no difference to the substantive outcome. In both Nulty and Graves the eventual decision was the same (in Graves, the Court of Appeal upheld the first instance decision that there was no negligence). The impression in both cases is that the Court of Appeal wanted to take the opportunity to get the law back on track.